Skip to contentSkip to conversation

x86, sched: fix up typo in topology detection

PATCH New Dave Hansen 2 participants 2 messages Sep 30, 2014 – Oct 3, 2014 x86

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 x86, sched: fix up typo in topology detection

From: Dave Hansen <[email protected]>

Commit cebf15eb09 (in linux-next now) added some code to try to detect the situation where we have a NUMA node inside of the "DIE" sched domain. It detected this by looking for cpus which match_die() but do not match NUMA nodes via topology_same_node().

I wrote it up as:

	if (match_die(c, o) == !topology_same_node(c, o))

which actually seemed to work some of the time, albiet accidentally.

It should have been doing an &&, not an ==.

This code essentially chopped off the "DIE" domain on one of Andrew Morton's systems. He reported that this patch fixed his issue.

---

 b/arch/x86/kernel/smpboot.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kernel/smpboot.c renamed from arch/x86/kernel/smpboot.c~fix-up-typo +1 -1
diff -puN arch/x86/kernel/smpboot.c~fix-up-typo arch/x86/kernel/smpboot.c
@@ -445,7 +445,7 @@ void set_cpu_sibling_map(int cpu)
 			} else if (i != cpu && !c->booted_cores)
 				c->booted_cores = cpu_data(i).booted_cores;
 		}
-		if (match_die(c, o) == !topology_same_node(c, o))
+		if (match_die(c, o) && !topology_same_node(c, o))
 			primarily_use_numa_for_topology();
 	}
 }

_

#1 lore raw

[tip:sched/core] sched/x86: Fix up typo in topology detection

sched/x86: Fix up typo in topology detection
Commit-ID:  728e5653e6fdb2a0892e94a600aef8c9a036c7eb
Gitweb:     http://git.kernel.org/tip/728e5653e6fdb2a0892e94a600aef8c9a036c7eb
Author:     Dave Hansen <[email protected]>
AuthorDate: Tue, 30 Sep 2014 14:45:46 -0700
Committer:  Ingo Molnar <[email protected]>
CommitDate: Fri, 3 Oct 2014 05:46:52 +0200

sched/x86: Fix up typo in topology detection

Commit:

cebf15eb09a2 ("x86, sched: Add new topology for multi-NUMA-node CPUs")

some code to try to detect the situation where we have a NUMA node inside of the "DIE" sched domain.

It detected this by looking for cpus which match_die() but do not match NUMA nodes via topology_same_node().

I wrote it up as:

	if (match_die(c, o) == !topology_same_node(c, o))

which actually seemed to work some of the time, albiet accidentally.

It should have been doing an &&, not an ==.

This code essentially chopped off the "DIE" domain on one of Andrew Morton's systems. He reported that this patch fixed his issue.

---
 arch/x86/kernel/smpboot.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kernel/smpboot.c +1 -1
index 8de8eb7..bae9e09 100644
@@ -445,7 +445,7 @@ void set_cpu_sibling_map(int cpu)
 			} else if (i != cpu && !c->booted_cores)
 				c->booted_cores = cpu_data(i).booted_cores;
 		}
-		if (match_die(c, o) == !topology_same_node(c, o))
+		if (match_die(c, o) && !topology_same_node(c, o))
 			primarily_use_numa_for_topology();
 	}
 }